def now(config):
df = get_health.now(config,
"br")[config["br"]["simulacovid"]["columns"]["cnes"]]
# merge cases
cases = get_cities_cases.now(config, "br")
cases = cases[cases["is_last"] == True].drop(config["br"]["cases"]["drop"],
1)
df = df.merge(cases, on="city_id", how="left", suffixes=("", "_y"))
df = df[[c for c in df.columns if not c.endswith("_y")]]
health_region_rate = (df[[
"health_region_id", "health_region_notification_rate"
]].dropna().groupby("health_region_id").mean())
# get notification for cities without cases
df["health_region_notification_rate"] = df.apply(
lambda row: _recover_notification_rate(row, health_region_rate),
axis=1)
df["notification_rate"] = np.where(
df["notification_rate"].isnull(),
df["health_region_notification_rate"],
df["notification_rate"],
)
df["last_updated"] = pd.to_datetime(df["last_updated"])
return df
def now(config):
df = get_health.now(config,
"br")[config["br"]["simulacovid"]["columns"]["cnes"]]
# merge cases
cases = get_cases.now(config, "br")
cases = cases[cases["is_last"] == True].drop(config["br"]["cases"]["drop"],
1)
df = df.merge(cases, on="city_id", how="left")
states_rate = (df[["state_id", "state_notification_rate"
]].dropna().groupby("state_id").mean())
# get notification for cities without cases
df["state_notification_rate"] = df.apply(
lambda row: _fix_state_notification(row, states_rate), axis=1)
df["notification_rate"] = np.where(
df["notification_rate"].isnull(),
df["state_notification_rate"],
df["notification_rate"],
)
df["last_updated"] = pd.to_datetime(df["last_updated"])
return df
def now(config):
# Get resource data
df = (get_health.now(config, "br").groupby([
"country_iso",
"country_name",
"state_num_id",
"state_id",
"state_name",
"health_region_id",
"health_region_name",
"last_updated_number_beds",
"author_number_beds",
"last_updated_number_icu_beds",
"author_number_icu_beds",
]).agg({
"population": sum,
"number_beds": sum,
"number_icu_beds": sum
}).reset_index().sort_values("health_region_id").set_index(
"health_region_id"))
df = get_situation_indicators(
df,
data=get_health_region_cases.now(config),
place_id="health_region_id",
rules=config["br"]["farolcovid"]["rules"],
classify="situation_classification",
)
df = get_control_indicators(
df,
data=get_health_region_rt.now(config),
place_id="health_region_id",
rules=config["br"]["farolcovid"]["rules"],
classify="control_classification",
)
df = get_trust_indicators(
df,
data=get_health_region_cases.now(config),
place_id="health_region_id",
rules=config["br"]["farolcovid"]["rules"],
classify="trust_classification",
)
df = get_capacity_indicators(
df,
place_id="health_region_id",
config=config,
rules=config["br"]["farolcovid"]["rules"],
classify="capacity_classification",
)
cols = [col for col in df.columns if "classification" in col]
df["overall_alert"] = df.apply(
lambda row: get_overall_alert(row[cols]),
axis=1) # .replace(config["br"]["farolcovid"]["categories"])
return df.reset_index()
def now(config, country="br"):
if country == "br":
infectious_period = (
config["br"]["seir_parameters"]["severe_duration"] +
config["br"]["seir_parameters"]["critical_duration"])
# Get data & clean table
df = (download_brasilio_table(config["br"]["cases"]["url"]).query(
"place_type == 'state'").fillna(0).rename(
columns=config["br"]["cases"]["rename"]).assign(
last_updated=lambda x: pd.to_datetime(x["last_updated"])).
sort_values(
["state_id", "last_updated"]).groupby("state_id").apply(
lambda group: get_until_last(group)).reset_index(
drop=True).drop(columns="estimated_population_2019"))
# Fix places_ids by city_id => Get health_region_id
df = df.merge(
get_places_id.now(config).assign(
state_num_id=lambda df: df["state_num_id"].astype(int), )[[
"state_name",
"state_id",
"state_num_id",
]].drop_duplicates(),
on="state_id",
)
cols = [
"state_num_id",
"state_id",
"state_name",
"last_updated",
"confirmed_cases",
"deaths",
"daily_cases",
"new_deaths",
]
# Add population data from CNES
df = df[cols].merge(
get_health.now(config).assign(
state_num_id=lambda df: df["state_num_id"].astype(int),
).groupby("state_num_id")["population"].sum().reset_index(),
on="state_num_id",
)
# Transform cases data
df = (
df.groupby([
"state_num_id",
"state_id",
"state_name",
]).apply(correct_negatives) # correct negative values
.pipe(
get_infectious_period_cases,
infectious_period,
config["br"]["cases"],
"state_num_id",
) # get infectious period cases
.rename(columns=config["br"]["cases"]["rename"]))
# Get indicators of mavg & growth
df = get_mavg_indicators(df, "daily_cases", place_id="state_num_id")
df = get_mavg_indicators(df, "new_deaths", place_id="state_num_id")
# Get notification rates & active cases on date
df = df.merge(
get_notification_rate.now(df, "state_num_id"),
on=["state_num_id", "last_updated"],
how="left",
).assign(active_cases=lambda x: np.where(
x["notification_rate"].isnull(),
np.nan, # round(x["infectious_period_cases"], 0),
round(x["infectious_period_cases"] / x["notification_rate"], 0),
))
return df
def now(config, country="br"):
if country == "br":
infectious_period = (
config["br"]["seir_parameters"]["severe_duration"] +
config["br"]["seir_parameters"]["critical_duration"])
# Filter and rename columns
df = (download_brasilio_table(
config["br"]["cases"]["url"]).query("place_type == 'city'").dropna(
subset=["city_ibge_code"]).fillna(0).rename(
columns=config["br"]["cases"]["rename"]).assign(
last_updated=lambda x: pd.to_datetime(x[
"last_updated"])).sort_values(
["city_id", "state_id",
"last_updated"]).groupby("city_id").
apply(lambda group: get_until_last(group)).reset_index(
drop=True).drop(
["city_name", "estimated_population_2019"],
1).assign(city_id=lambda df: df["city_id"].astype(int)))
# Fix places name & ID + Get total population
places_ids = get_health.now(config).assign(
city_id=lambda df: df["city_id"].astype(int))
df = df.merge(
places_ids[[
"city_id",
"city_name",
"health_region_name",
"health_region_id",
"state_name",
"state_num_id",
"population",
]],
on="city_id",
)
# Correct negative values, get infectious period cases and get median of new cases
df = (df.groupby("city_id").apply(correct_negatives).pipe(
get_infectious_period_cases,
infectious_period,
config["br"]["cases"],
"city_id",
).rename(columns=config["br"]["cases"]["rename"]))
# Get indicators of mavg & growth
df = get_mavg_indicators(df, "daily_cases", place_id="city_id")
df = get_mavg_indicators(df, "new_deaths", place_id="city_id")
# Get notification rates & active cases on date
df = df.merge(
get_notification_rate.now(df, "health_region_id"),
on=["health_region_id", "last_updated"],
how="left",
).assign(
active_cases=lambda x: np.where(
x["notification_rate"].isnull(),
np.nan, # round(x["infectious_period_cases"], 0),
round(x["infectious_period_cases"] / x["notification_rate"], 0
),
),
city_id=lambda x: x["city_id"].astype(int),
)
return df